Restoration in Mediterranean-climate grasslands is strongly impeded by lack of native propagules and competition with exotic grasses and forbs. We report on a study testing several methods for exotic plant control combined with planting native grasses to restore prairies in former agricultural land in coastal California. Specifically we compared tarping (shading out recently germinated seedlings with black plastic) once, tarping twice, topsoil removal, herbicide (glyphosate), and a control treatment in factorial combinations with or without wood mulch. Into each treatment we planted three native grass species (Elymus glaucus, Hordeum brachyantherum, and Stipa pulchra) and monitored plant survival and cover for three growing seasons. Survival of native grass species was high in all treatments, but was slightly lower in unmulched soil removal and control treatments in the first 2 yr. Mulching, tarping, and herbicide were all effective in reducing exotic grass cover and enhancing native grass cover for the first 2 yr, but by the third growing season cover of the plant guilds and bare ground had mostly converged, primarily because of the declining effects of the initial treatments. Mulching and tarping were both considerably more expensive than herbicide treatment. Topsoil removal was less effective in increasing native grass cover likely because soil removal altered the surface hydrology in this system. Our results show that several treatments were effective in enhancing native grass establishment, but that longer term monitoring is needed to evaluate the efficacy of restoration efforts. The most appropriate approach to controlling exotics to restore specific grassland sites will depend not only on the effectiveness, but also on relative costs and site constraints.